CoolFace
Apppublic

elliottsax/autocoder-test-planning-worker

sourceHugging Facemitupdated 9mo agoView on Hugging Face
0likes
App README

Autocoder Test Planning Worker 🧪

24/7 Autonomous Test Planning Worker

This Hugging Face Space runs continuous test planning analysis, discovering test requirements, edge cases, and failure modes in real-time.

Features

  • —✅ Continuous Analysis - Runs 24/7 automatically
  • —✅ 15 Perspectives - Analyzes from 15 different testing angles
  • —✅ Cloud Storage - Results pushed to HF Datasets
  • —✅ Real-time Monitoring - Live status dashboard
  • —✅ Free Tier Compatible - Runs on HF free infrastructure

How It Works

  1. 1.Analysis Loop - Worker runs continuous test planning iterations
  2. 2.Perspective Rotation - Cycles through 15 different analysis perspectives
  3. 3.Result Storage - Pushes test specifications to HF Dataset
  4. 4.Monitoring - UI updates every 5 seconds

Analysis Perspectives

  1. 1.What Are We Testing? - Scope and boundaries
  2. 2.How Can It Fail? - Failure mode analysis
  3. 3.Integration Points - Where systems connect
  4. 4.Edge Cases - Boundary conditions
  5. 5.Performance - Speed and efficiency
  6. 6.Backwards Compatibility - Breaking changes
  7. 7.Data Integrity - Data correctness
  8. 8.Error Handling - Failure recovery
  9. 9.User Experience - UX implications
  10. 10.Security - Vulnerability analysis
  11. 11.Test Coverage - Coverage gaps
  12. 12.Test Pyramid - Test structure
  13. 13.Mocking Strategy - Test isolation
  14. 14.CI/CD Integration - Automation
  15. 15.Big Picture Review - Holistic analysis

Configuration

Required Secrets

Add these in Space Settings → Repository Secrets:

  • —HF_TOKEN - Your Hugging Face access token (write permission)
  • —HF_USERNAME - Your HF username (optional, defaults to 'autocoder')

Dataset Output

Results are saved to: {HF_USERNAME}/autocoder-test-planning-results

Usage

Via UI

  1. 1.Visit this Space
  2. 2.View real-time status
  3. 3.Start/stop worker as needed
  4. 4.View test planning logs

Programmatically

python
from datasets import load_dataset

# Load all test planning results
dataset = load_dataset("autocoder/autocoder-test-planning-results")

# Get latest results
latest = dataset["train"][-1]
print(latest)

Integration with Other Workers

This worker's output is used by:

  • —Analysis Worker (provides code insights)
  • —Implementation Worker (uses test plans to guide coding)
  • —Monitoring Dashboard (visualizes test coverage)

Local Development

bash
# Install dependencies
pip install -r requirements.txt

# Set environment variables
export HF_TOKEN="your_token"
export HF_USERNAME="your_username"

# Run locally
python app.py

Deployment

  1. 1.Fork this Space or create new
  2. 2.Configure secrets
  3. 3.Space will auto-start worker

Monitoring

  • —Status Updates: Every 5 seconds
  • —Test Planning Iterations: Every 60 seconds
  • —Dataset Uploads: After each iteration

Test Specifications Generated

Each iteration produces:

  • —Failure mode analysis
  • —Integration point mapping
  • —Edge case catalog
  • —Test specifications with code
  • —Coverage analysis
  • —Risk assessment

Cost

  • —Free Tier: Fully compatible
  • —Persistent: Space keeps running 24/7
  • —Storage: Dataset grows over time (monitor quota)

Support

For issues or questions, open an issue in the main repository.

License

MIT License - See LICENSE file